diff --git a/llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll b/llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll
new file mode 100644
index 0000000000000..9efe97004fd92
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll
@@ -0,0 +1,188 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple aarch64-none-linux-gnu < %s | FileCheck %s
+
+; Function Attrs: nofree norecurse nosync nounwind memory(argmem: readwrite) uwtable
+define dso_local void @foo(i32 noundef %limit, ptr nocapture noundef %out, ptr nocapture noundef readonly %y) local_unnamed_addr #0 {
+; CHECK-LABEL: foo:
+; CHECK:       // %bb.0: // %entry
+; CHECK-NEXT:    // kill: def $w0 killed $w0 def $x0
+; CHECK-NEXT:    cmp w0, #1
+; CHECK-NEXT:    b.lt .LBB0_10
+; CHECK-NEXT:  // %bb.1: // %for.cond1.preheader.us.preheader
+; CHECK-NEXT:    mov w10, w0
+; CHECK-NEXT:    mov x8, xzr
+; CHECK-NEXT:    mov x9, xzr
+; CHECK-NEXT:    and x11, x10, #0xfffffff0
+; CHECK-NEXT:    add x12, x1, #32
+; CHECK-NEXT:    ubfiz x13, x0, #2, #32
+; CHECK-NEXT:    add x14, x2, #16
+; CHECK-NEXT:    b .LBB0_3
+; CHECK-NEXT:    .p2align 5, , 16
+; CHECK-NEXT:  .LBB0_2: // %for.cond1.for.cond.cleanup3_crit_edge.us
+; CHECK-NEXT:    // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    add x9, x9, #1
+; CHECK-NEXT:    add x12, x12, x13
+; CHECK-NEXT:    add x8, x8, x10
+; CHECK-NEXT:    cmp x9, x10
+; CHECK-NEXT:    b.eq .LBB0_10
+; CHECK-NEXT:  .LBB0_3: // %for.cond1.preheader.us
+; CHECK-NEXT:    // =>This Loop Header: Depth=1
+; CHECK-NEXT:    // Child Loop BB0_6 Depth 2
+; CHECK-NEXT:    // Child Loop BB0_9 Depth 2
+; CHECK-NEXT:    ldrsh w15, [x2, x9, lsl #1]
+; CHECK-NEXT:    cmp w0, #16
+; CHECK-NEXT:    b.hs .LBB0_5
+; CHECK-NEXT:  // %bb.4: // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    mov x18, xzr
+; CHECK-NEXT:    b .LBB0_8
+; CHECK-NEXT:    .p2align 5, , 16
+; CHECK-NEXT:  .LBB0_5: // %vector.ph
+; CHECK-NEXT:    // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    mov x16, x14
+; CHECK-NEXT:    mov x17, x12
+; CHECK-NEXT:    mov x18, x11
+; CHECK-NEXT:    .p2align 5, , 16
+; CHECK-NEXT:  .LBB0_6: // %vector.body
+; CHECK-NEXT:    // Parent Loop BB0_3 Depth=1
+; CHECK-NEXT:    // => This Inner Loop Header: Depth=2
+; CHECK-NEXT:    dup v0.8h, w15
+; CHECK-NEXT:    subs x18, x18, #16
+; CHECK-NEXT:    ldp q1, q2, [x16, #-16]
+; CHECK-NEXT:    add x16, x16, #32
+; CHECK-NEXT:    ldp q4, q3, [x17, #-32]
+; CHECK-NEXT:    smlal v4.4s, v0.4h, v1.4h
+; CHECK-NEXT:    smlal2 v3.4s, v0.8h, v1.8h
+; CHECK-NEXT:    ldp q6, q5, [x17]
+; CHECK-NEXT:    smlal v6.4s, v0.4h, v2.4h
+; CHECK-NEXT:    smlal2 v5.4s, v0.8h, v2.8h
+; CHECK-NEXT:    stp q4, q3, [x17, #-32]
+; CHECK-NEXT:    stp q6, q5, [x17], #64
+; CHECK-NEXT:    b.ne .LBB0_6
+; CHECK-NEXT:  // %bb.7: // %middle.block
+; CHECK-NEXT:    // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    cmp x11, x10
+; CHECK-NEXT:    mov x18, x11
+; CHECK-NEXT:    b.eq .LBB0_2
+; CHECK-NEXT:  .LBB0_8: // %for.body4.us.preheader
+; CHECK-NEXT:    // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    add x17, x18, x8
+; CHECK-NEXT:    sub x16, x10, x18
+; CHECK-NEXT:    add x18, x2, x18, lsl #1
+; CHECK-NEXT:    add x17, x1, x17, lsl #2
+; CHECK-NEXT:    .p2align 5, , 16
+; CHECK-NEXT:  .LBB0_9: // %for.body4.us
+; CHECK-NEXT:    // Parent Loop BB0_3 Depth=1
+; CHECK-NEXT:    // => This Inner Loop Header: Depth=2
+; CHECK-NEXT:    ldrsh w3, [x18], #2
+; CHECK-NEXT:    subs x16, x16, #1
+; CHECK-NEXT:    ldr w4, [x17]
+; CHECK-NEXT:    madd w3, w3, w15, w4
+; CHECK-NEXT:    str w3, [x17], #4
+; CHECK-NEXT:    b.ne .LBB0_9
+; CHECK-NEXT:    b .LBB0_2
+; CHECK-NEXT:  .LBB0_10: // %for.cond.cleanup
+; CHECK-NEXT:    ret
+entry:
+  %cmp26 = icmp sgt i32 %limit, 0
+  br i1 %cmp26, label %for.cond1.preheader.us.preheader, label %for.cond.cleanup
+
+for.cond1.preheader.us.preheader:                 ; preds = %entry
+  %0 = zext i32 %limit to i64
+  %wide.trip.count34 = zext i32 %limit to i64
+  %min.iters.check = icmp ult i32 %limit, 16
+  %n.vec = and i64 %wide.trip.count34, 4294967280
+  %cmp.n = icmp eq i64 %n.vec, %wide.trip.count34
+  br label %for.cond1.preheader.us
+
+for.cond1.preheader.us:                           ; preds = %for.cond1.preheader.us.preheader, %for.cond1.for.cond.cleanup3_crit_edge.us
+  %indvars.iv30 = phi i64 [ 0, %for.cond1.preheader.us.preheader ], [ %indvars.iv.next31, %for.cond1.for.cond.cleanup3_crit_edge.us ]
+  %arrayidx.us = getelementptr inbounds i16, ptr %y, i64 %indvars.iv30
+  %1 = load i16, ptr %arrayidx.us, align 2, !tbaa !6
+  %conv.us = sext i16 %1 to i32
+  %2 = mul nsw i64 %indvars.iv30, %0
+  br i1 %min.iters.check, label %for.body4.us.preheader, label %vector.ph
+
+vector.ph:                                        ; preds = %for.cond1.preheader.us
+  %broadcast.splatinsert = insertelement <8 x i32> poison, i32 %conv.us, i64 0
+  %broadcast.splat = shufflevector <8 x i32> %broadcast.splatinsert, <8 x i32> poison, <8 x i32> zeroinitializer
+  %broadcast.splatinsert37 = insertelement <8 x i32> poison, i32 %conv.us, i64 0
+  %broadcast.splat38 = shufflevector <8 x i32> %broadcast.splatinsert37, <8 x i32> poison, <8 x i32> zeroinitializer
+  br label %vector.body
+
+vector.body:                                      ; preds = %vector.body, %vector.ph
+  %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
+  %3 = getelementptr inbounds i16, ptr %y, i64 %index
+  %wide.load = load <8 x i16>, ptr %3, align 2, !tbaa !6
+  %4 = getelementptr inbounds i16, ptr %3, i64 8
+  %wide.load36 = load <8 x i16>, ptr %4, align 2, !tbaa !6
+  %5 = sext <8 x i16> %wide.load to <8 x i32>
+  %6 = sext <8 x i16> %wide.load36 to <8 x i32>
+  %7 = mul nsw <8 x i32> %broadcast.splat, %5
+  %8 = mul nsw <8 x i32> %broadcast.splat38, %6
+  %9 = add nuw nsw i64 %index, %2
+  %10 = getelementptr inbounds i32, ptr %out, i64 %9
+  %wide.load39 = load <8 x i32>, ptr %10, align 4, !tbaa !10
+  %11 = getelementptr inbounds i32, ptr %10, i64 8
+  %wide.load40 = load <8 x i32>, ptr %11, align 4, !tbaa !10
+  %12 = add nsw <8 x i32> %7, %wide.load39
+  %13 = add nsw <8 x i32> %8, %wide.load40
+  store <8 x i32> %12, ptr %10, align 4, !tbaa !10
+  store <8 x i32> %13, ptr %11, align 4, !tbaa !10
+  %index.next = add nuw i64 %index, 16
+  %14 = icmp eq i64 %index.next, %n.vec
+  br i1 %14, label %middle.block, label %vector.body, !llvm.loop !12
+
+middle.block:                                     ; preds = %vector.body
+  br i1 %cmp.n, label %for.cond1.for.cond.cleanup3_crit_edge.us, label %for.body4.us.preheader
+
+for.body4.us.preheader:                           ; preds = %for.cond1.preheader.us, %middle.block
+  %indvars.iv.ph = phi i64 [ 0, %for.cond1.preheader.us ], [ %n.vec, %middle.block ]
+  br label %for.body4.us
+
+for.body4.us:                                     ; preds = %for.body4.us.preheader, %for.body4.us
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body4.us ], [ %indvars.iv.ph, %for.body4.us.preheader ]
+  %arrayidx6.us = getelementptr inbounds i16, ptr %y, i64 %indvars.iv
+  %15 = load i16, ptr %arrayidx6.us, align 2, !tbaa !6
+  %conv7.us = sext i16 %15 to i32
+  %mul.us = mul nsw i32 %conv7.us, %conv.us
+  %16 = add nuw nsw i64 %indvars.iv, %2
+  %arrayidx10.us = getelementptr inbounds i32, ptr %out, i64 %16
+  %17 = load i32, ptr %arrayidx10.us, align 4, !tbaa !10
+  %add11.us = add nsw i32 %mul.us, %17
+  store i32 %add11.us, ptr %arrayidx10.us, align 4, !tbaa !10
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond.not = icmp eq i64 %indvars.iv.next, %wide.trip.count34
+  br i1 %exitcond.not, label %for.cond1.for.cond.cleanup3_crit_edge.us, label %for.body4.us, !llvm.loop !16
+
+for.cond1.for.cond.cleanup3_crit_edge.us:         ; preds = %for.body4.us, %middle.block
+  %indvars.iv.next31 = add nuw nsw i64 %indvars.iv30, 1
+  %exitcond35.not = icmp eq i64 %indvars.iv.next31, %wide.trip.count34
+  br i1 %exitcond35.not, label %for.cond.cleanup, label %for.cond1.preheader.us, !llvm.loop !17
+
+for.cond.cleanup:                                 ; preds = %for.cond1.for.cond.cleanup3_crit_edge.us, %entry
+  ret void
+}
+
+attributes #0 = { nofree norecurse nosync nounwind memory(argmem: readwrite) uwtable "frame-pointer"="non-leaf" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="neoverse-n1" "target-features"="+aes,+crc,+dotprod,+fp-armv8,+fullfp16,+lse,+neon,+outline-atomics,+ras,+rcpc,+rdm,+sha2,+spe,+ssbs,+v8.1a,+v8.2a,+v8a,-fmv" }
+
+!llvm.module.flags = !{!0, !1, !2, !3, !4}
+!llvm.ident = !{!5}
+
+!0 = !{i32 1, !"wchar_size", i32 4}
+!1 = !{i32 8, !"PIC Level", i32 2}
+!2 = !{i32 7, !"PIE Level", i32 2}
+!3 = !{i32 7, !"uwtable", i32 2}
+!4 = !{i32 7, !"frame-pointer", i32 1}
+!5 = !{!"clang version 17.0.0 (ssh://eu-gerrit-1.euhpc.arm.com:29418/dsg-dps/mirrors/llvm 2b55808b298965d2d49727b291198d4e2551d24a)"}
+!6 = !{!7, !7, i64 0}
+!7 = !{!"short", !8, i64 0}
+!8 = !{!"omnipotent char", !9, i64 0}
+!9 = !{!"Simple C/C++ TBAA"}
+!10 = !{!11, !11, i64 0}
+!11 = !{!"int", !8, i64 0}
+!12 = distinct !{!12, !13, !14, !15}
+!13 = !{!"llvm.loop.mustprogress"}
+!14 = !{!"llvm.loop.isvectorized", i32 1}
+!15 = !{!"llvm.loop.unroll.runtime.disable"}
+!16 = distinct !{!16, !13, !15, !14}
+!17 = distinct !{!17, !13}